Speaking the right media language[1]

The Honourable Justice S Rares[2]

Justice Rares discusses issues arising from the prevalence of social media, including how should courts and tribunals deal with social media issues that arise from media and parties using social media in and around hearings, and, how should judicial officers and tribunal members deal with social media issues in their personal lives, especially when those interact with their official roles.

Australian judges, at least, in their day jobs, are not generally known for either their expertise or engagement with social media, or popular culture. Of course, this generalisation does not apply to the Honouable Michael Kirby AC, or the moonlighting novelist, the Honourable Ian Callinan AC or the many serving and retired judges who conduct Royal Commissions. Indeed, Australian judges, and I dare say tribunal members, appear much more shy than some of their counterparts overseas. It is hard to imagine the justices of the High Court appearing as judges on Masterchef, as Baroness Hale, the President of the Supreme Court of the United Kingdom, did in April 2018.[3] Even so, things have moved on since 2012 when her predecessor, then Master of the Rolls, Lord Neuberger, criticised the two Supreme Court justices and three Lords Justices of Appeal who had participated in judging on an episode of the show what members of the bench and bar cooked.

Generally, as I will explain later, Australian judges have trodden warily in social and mass media spaces. Few of us are well known figures, in contrast to some of our counterparts in the United States of America.

You may be aware that this year, two biopics, “RBG” and “On the Basis of Sex”, have been released about the life of the well-known liberal United States Supreme Court Justice, Ruth Bader Ginsburg. She was appointed by President Bill Clinton in 1993, and since then her Honor has developed a large online presence on social media. Never a shrinking violet, in July 2016, Justice Ginsburg entered the Presidential election campaign in an interview with a CNN journalist (Joan Biskupic) who was researching a book on Chief Justice John Roberts. RBG, as she is known, described then Republican challenger to Hillary Clinton, Donald Trump, as:[4]

a faker. He has no consistency about him. He says whatever comes into his head at the moment. He really has an ego…

On 13 July 2016, those remarks got her into their subject’s favourite social media platform, Twitter, as follows:

As the copy paste of that tweet shows, within hours of its release over 30,000 people had indicated that they liked Mr Trump’s response. That figure belies how many others read it, to say nothing of any who disliked it. The account’s controller is not known to disseminate views of his opponents — what they might have had to say would be just dismissed as “fake news”.

Apparently younger people appear almost as frantic as the “Tweeter-in-Chief” to share quotes, images, and “memes” of Justice Ginsburg online. She has been called a “Social Media Darling”, and is easily one of the most recognisable judges in America, especially among those members of the public who do not work in the legal sector.[5]

The character people ascribe to Justice Ginsburg on social media may not necessarily accord with reality, but that does not stop that character being shared and discussed by millions. One popular meme refers to her Honor as the “Notorious RBG”, being what my associate informs me is a pun on someone known as the late American rapper “Notorious BIG”. Video clips online feature actor Kate McKinnon, from “Saturday Night Live”, portraying the real life judge dropping “Gins-burns”. The clips are shared by fans after Justice Ginsberg’s significant opinions are published. In one of the recent films, she appears lifting weights and wearing a “Super Diva” sweat-shirt. Her Honor is reputed to have t-shirts made, or sent to her, that reference current popular memes, which she gives out to friends.

The 86-year old might seem to Australian eyes an unlikely judicial lightning rod for the social media generation. The online fascination with her perhaps exemplifies the unusual nature of social media and raises some of the issues which can arise from publications in the cyber world. Indeed, Mr Trump’s robust tweet above is a vivid illustration of what social media interactions can involve or unleash.

Of course, there is nothing new in publishers having to face up to the consequences from their expressing themselves publicly. The law of defamation knows no cyber boundaries. In November 1904, the sadly now defunct satirical magazine Punch published a book review that began as follows:[6]


Extract from the Recess Diary of Toby, MP.

Been reading ‘Fifty Years of Fleet Street’ just issued by Macmillan. Purports to be the ‘Life and Recollections of Sir John Robinson’, the man who made, and for a quarter of a century maintained at high level, the Daily News. The story is written by Mr F M Thomas, who has added a new terror to death.

It might be said that social media has added a new terror to life. The social commentator, Bernard Salt, recently wrote that he suspected that the real audience for social media invective is not the cited target. Rather, he thought that the invective’s real purpose was to secure the support of persons who followed its author.[7]

In today’s age, courts and tribunals must accept and adapt to the reality that social media is now an established societal institution. That means that, whether or not a judicial or statutory tribunal officer uses social media themselves, his or her public, and sometimes personal work, actions, images and words, will appear there. How then, can they learn to speak the right social media language?

The environment in which courts and tribunals work involves parties who appear before us using social media to help, promote, or, on occasion, unwittingly, hinder their causes, or to explain why something did not turn out as well for them as their earlier online predictions had presaged. Frequently now, media and members of the public comment in real time on proceedings regardless of whether the communicator has any personal knowledge of what is going on. Perhaps this is not new but it is now more pervasive and accessible than in days past. Sometimes, in cyber space exchanges, knowledge of what one is talking about can be a hindrance or dismissed as an “alternative fact”. Ignorance may be bliss, but online but may be highly infectious.

Over the last 15 years, many of the present cohort of judicial officers and tribunal members have had one or more social media presences of their own, such as on Facebook, WeChat and Twitter, including from a time before their appointment. Many have retained those accounts, and some have initiated a social media presence since their appointment, as they are entitled to do. However, there are dangers to this activity.

In this paper, I want to discuss some increasingly common issues arising from the prevalence in our lives of social media. These include how should courts and tribunals deal with social media issues that arise from media and parties using social media in and around hearings themselves, and, how should judicial officers and tribunal members deal with social media issues in their personal lives, especially when those interact with their official roles.


The expression “social media” refers to communication involving a kind of social interaction by technological means. It includes website and mobile phone applications that allow people to communicate or express thoughts or images unilaterally or interactively with the world at large. An interaction or communication on social media is generally instant, can be wide reaching, and will often be one-sided, in that its author does not call for, or rely on, a response. When a response does come, it may be a reaction rather than a direct response. Someone viewing such a communication might acknowledge his or her support, approval or disavowal of the sentiments that the original poster made available to the world at large by using a “like”, “share” or “dislike” icon on the social media. I am reliably informed that, perhaps in order to avoid directness, the “dislike” icon is more generally an “angry” or a “sad” face icon.

The most popular social media platforms today are Facebook, Youtube, Instagram, WhatsApp, Snapchat, WeChat and Twitter. They each are open to anyone to join, and reach slightly different (though potentially overlapping) audiences. On Facebook and Instagram, users (at least individuals, as opposed to corporations that also use the platforms) generally post content that they wish to be broadly available to their “friends”. Many of those persons are people who have connected with the instigator of the communication because their only awareness or connection with him or her is through the platform. Others either know the instigator or genuinely are interested to learn what he or she has to say and share. Youtube and Twitter generally make videos and posts available to the public at large. The public can view a video or post, share it among their own network of contacts, or comment on it, including (if the instigator has allowed this) to engage in a discussion with that instigator and or other users of the platform at large. On WhatsApp, WeChat and Snapchat, material is generally shared one-on-one or within small groups of users.

Despite the different reaches of each of these social media platforms, and the many others that exist for different niches, there remain risks and issues for judicial officers and tribunal members in trying or failing to “speak the language” of social media effectively.

Risks and issues arise because, it is fair to say, the use of social media in today’s society is pervasive. In 2019, one source has precisely identified that there are 3.48 billion “social media users”, who, if all were individuals, make up about 80% of all internet users, and almost 50% of the global population.[8] I am not one of them. In Australia, as at January 2019, of a population of about 25 million, there are 15 million Facebook accounts, 9 million Instagram accounts and 4.7 million Twitter accounts.[9] Of course, not all of these accounts belong to individuals since communication on social media platforms is nowadays almost an essential feature of businesses of all sizes, from multinational corporations to one’s local café, to say nothing of their use by other organisations such as political parties, at least when they do so overtly. Another source asserts that 15 million Australians visit YouTube each month, and 2.9 million use WeChat, a China-based social-media-cum-messaging platform.[10]

Depending on the platform, social media can afford each of those users a potentially vast audience. And the immediate audience can grow exponentially as posts are shared by other users. A private post that the author intended to be made only to his or her platform friends, can be reposted, retweeted or sent so that, eventually, it might be read by millions. This can occur without the help, and even against the wishes of the person who made the original post. Once shared beyond an individual’s private network, a post generally can never be recalled. Even those shared privately and kept private may evade deletion. As every employer now knows, the best way to find out about an actual or potential employee is to search for them on social media.

This matters for courts and tribunals for main three reasons. First, a judicial officer’s or tribunal member’s use of social media can give rise to an apprehension of bias because of what he or she puts online about himself or herself.

Second, it may be difficult, especially for less digitally adept judicial or tribunal officers to control who can have access to one’s social media account and who can republish a post that the judge or member has made. Your online friends may be very fairweather ones, or even wolves in sheep’s clothing. For example, a person who has 200 online “friends”, each of whom has his or her own 100 additional unique online friends, would be within the reach of 20,000 people when sharing a post. In social media terms, terrifyingly, that extended circle of one’s not very intimate “friends” is possibly a modest collective. There is obviously a risk that potentially a litigant, or his, her or its counsel or lawyers, will be among that audience and may perceive something in what they see online that could cause a problem for the judge or member hearing their matter.

Third, information about what is happening in a court or tribunal room in Melbourne or Sydney, Ballarat or Goulburn, can have a global reach in an instant. In high profile cases, it is increasingly difficult in open court to control the accidental disclosure, via social media, of confidential information or, in a jury trial, the subject-matter of a hearing in the absence of the jury, involving matters that may prejudice the proper conduct of the trial. This has become the more concerning with the demise of trained court reporters for traditional mass media outlets who understood what could and could not be reported, unlike a casual blogger sitting in a hearing who constantly is updating his or her blog with the latest utterances.

Part I: Apprehension of bias

In many aspects of daily life, there is a real danger that a judicial officer or tribunal member may create a situation where a party before him or her can raise a claim that the judge or member is actually or apparently biased. Often, social media engagements, for example those limited to 140 (or, now, 280) characters leave little room for subtlety or refinement. And once posted, a tweet or Facebook comment can be as indelible as the printed word. But, social media posts can be disseminated very widely and out of any context. Sometimes, in every-day interactions, such as personal face-to-face communications, we say things that are unguarded or casual without necessarily meaning them or thinking through their implications, which, on reflection, we did not really believe or we come to realise were wrong or just silly. Often, someone in conversation points this out and everyone forgets about it straight away. Not so with a post.

An administrative decision-maker will be found to have given rise to an apprehension of bias if a fair-minded lay person might think that the decision-maker might not bring a fair and impartial mind to the making of the decision. The hypothetical lay person is an objective observer of the proceedings and will be assumed to be properly informed as to their nature, the matters in issue and the conduct complained of.[11]

The mere fact that a decision-maker has previously expressed a view on the same or a similar subject does not, of itself, give rise to an apprehension that he or she will not bring a fair and impartial mind to the new decision to be made.[12] After all, decision-makers can be expected to apply the law and relevant policies in a consistent and predictable way. Likewise, decision-makers in the position of a tribunal member or administrative official frequently will have to decide the same issues raised by different persons in separate applications including when a number of persons make generic claims. A decision-maker must have a fair and unprejudiced mind when he or she comes to decide a question including one concerning a generic claim that he or she has addressed on another occasion. However, that does not mean that he or she must have a blank or empty mind on the topic. As Barwick CJ, McTiernan, Kitto, Taylor, Menzies, Windeyer and Owen JJ said in The Queen v Commonwealth Conciliation and Arbitration Commission; Ex p Angliss Group:[13]

Such a mind is not necessarily a mind which has not given thought to the subject matter or one which, having thought about it, has not formed any views or inclination of mind upon or with respect to it.

And, in Minister for Immigration and Multicultural Affairs v Jia Legeng,[14] Gleeson CJ and Gummow J said:

The state of mind described as bias in the form of prejudgment is one so committed to a conclusion already formed as to be incapable of alteration, whatever evidence or arguments may be presented. Natural justice does not require the absence of any predisposition or inclination for or against an argument or conclusion.

Chief Justice Gleeson and Gummow J recognised, in agreeing with the reasons of Hayne J, that the concept of apprehended bias, as an aspect of procedural fairness, has to accommodate, and may vary in, different decision-making environments.[15] Justice Hayne explained that the genesis of rules about the concept of judicial prejudgment is different from that of prejudgment in administrative contexts and that a range of considerations and differing consequences will arise depending on the source and context of the executive power being exercised.[16] His Honour recognised that specialised tribunals, such as the Refugee Review Tribunal, now the Administrative Appeals Tribunal, would “bring to the task of deciding an individual’s application a great deal of information and ideas which have been accumulated or formed in the course of deciding other applications”.[17] He said that such a decision-maker was expected to build up “expertise” in matters such as country information, saying:[18]

Often information of that kind is critical in deciding the fate of an individual’s application, but it is not suggested that to take it into account amounts to a want of procedural fairness by reason of prejudgment.

Justice Hayne elaborated by explaining that at least four distinct elements require consideration in examining an assertion that a decision-maker has prejudged or will prejudge, or that there is a real likelihood that a reasonable observer might reach such a conclusion. He said that the assertion of apprehended bias contains contentions that, first, the decision-maker has an opinion on a relevant aspect of a matter in issue in the particular case, second, he or she will apply that opinion to the matter in the case and, third, he or she:[19]

will do so without giving the matter fresh consideration in the light of whatever may be the facts and arguments relevant to the particular case. Most importantly, [fourthly] there is the assumption that the question which is said to have been prejudged is one which should be considered afresh in relation to the particular case.

Often enough, allegations of actual bias through prejudgment have been held to fail at the third of the steps I have identified. In 1894, it was said that:[20]

“preconceived opinions — though it is unfortunate that a judge should have any — do not constitute such a bias, nor even the expression of such opinions, for it does not follow that the evidence will be disregarded.” [Emphasis added.]

Allegations of apprehended bias through prejudgment are often dealt with similarly.[21]

Accordingly, the way in which a decision-maker may properly go about his or her task and what kind or degree of neutrality, if any, is to be expected of him or her will be relevant considerations in evaluating how and in what way the rules relating to apprehension of bias will be applied in a particular situation.[22] And, as his Honour concluded:[23]

Once it is recognised that there are elements of the decision-making process about which a decision-maker may legitimately form and hold views before coming to consider the exercise of a power in a particular case, it is evident that the area within which questions of actual or apprehended bias by prejudgment may arise is reduced accordingly.

A professional career can be ruined by a careless or foolish tweet or post. One example of the disastrous effect such a communication can have is what happened to a public relations executive, Justine Sacco. She posted the following tweet to her 170 Twitter followers before she left London on a flight from New York to holiday in South Africa:

By the time her plane touched down in Cape Town, her tweet had gone viral and been retweeted tens of thousand of times. A photographer snapped her in the arrivals hall. The next day, she was sacked from her employment. Her tweet created international outrage. She was denounced by the anonymous purveyors of social media correctness. Yet, when you think about what she did, it was to write a clumsy, silly joke. Had she said it in conversation with some friends, one or more may have said something discouraging, or she may have experienced an awkward silence or look. That would have made her realise that her comment was inappropriate. Or, her listeners may just have had a sense of humour, rather than outrage, and laughed.

Whatever may have been the outcome of a person to person situation, the use of social media and the retweeting with critical, if not denunciatory, comments created a virtual world lynch mob in which Ms Sacco could not be even perceived as having any amount of human frailty or a poor, but not malicious, sense of humour. Rather, she was vilified and became, for a time unemployable. And that was just because she wrote something silly as opposed to having, for example, constructed a deliberate campaign slogan for an extremist organisation such as the Ku Klux Klan.[24]

Things we used to say verbally as part of human communication to a small number of people can now become accessible, if said on social media, to millions instantaneously and in a permanent form. And, just as with direct face-to-face contact, posts can be misunderstood. Nor will the click of the delete button effectively erase an ill-considered post — there is the earlier retweet or republication, not to mention the Wayback Machine to ensure that you cannot hide your deletions in the virtual world.

So while judges and tribunal members should not avoid social media, they would be well advised, just as in their work, to consider carefully their choice of words and, when they publish, have regard to professional and ethical obligations as well as to common sense. In my generation, one commonly used test was to consider what would happen if what you were intending to write appeared on the front page of a daily newspaper — frequently it was a sobering idea.

Part II: Social media in the personal life of judges and tribunal members

A claim for apprehension of bias could possibly also arise if a member of the family or staff of a judge or tribunal member publishes something that reasonably could be attributed as being what the judge or member has said about a matter or issue before him or her. Thus, social media, like Pandora’s box, offers all manner of calamity. There is a risk that the online publication or revelation of a judge’s or tribunal member’s online posts, friends or interests might cause a reasonable layperson to consider that the judge or member might not decide a matter impartially.

Facebook “friends”

Often after giving a decision in a matter one might feel that friends are few and far between. That should not encourage too ready a disposition to “friend” people online.

One example of online stupidity occurred in Wisconsin in 2016. There, Judge Bitney was hearing a child custody dispute. He received a friend request from the mother soon after the close of argument, but before the resolution of the case.[25] You might think this is fanciful, but the judge accepted that request. Needless to say, he did not disclose the request or its acceptance to the father. A central issue in the case was the credibility of his Honour’s new friend’s allegations that the father had engaged in domestic abuse towards her. Ultimately, the judge decided the case in the mother’s favour. In the meantime, before he decided the case, the mother liked 18 of the judge’s posts about other aspects of his life. He did not respond to any of the mother’s communications in that period. It is important to understand that judges in Wisconsin are elected, as they are in 38 other States in the United States of America.

Later, on the day of the decision, the guardian ad litem uncovered the Facebook “friendship” when visiting the mother’s Facebook page to read a post made about the outcome of the proceedings. Subsequently, the judge denied that he had “liked” or commented on any posts made by the mother, but he did not deny that he had seen and read her posts before he gave his decision. The mother had made a large number of posts about domestic violence, supporting causes that aimed to prevent it.

The father then applied to the judge to quash his decision, to disqualify himself, and for the matter to be re-heard. Judge Bitney dismissed that application. On appeal, the Wisconsin Court of Appeals reversed that decision. The Court of Appeals said:[26]

First, the time when Judge Bitney and Carroll became Facebook “friends” would cause a reasonable person to question the judge’s partiality. Although Judge Bitney apparently had thousands of Facebook “friends”, Carroll was not simply one of the many people who “friended” him prior to this litigation. Rather, Carroll was a current litigant who reached out to Judge Bitney and requested to become his Facebook “friend” after a contested hearing, at which Judge Bitney was the sole decision-maker. Judge Bitney then took the affirmative step to accept this “friend” request before he issued his decision in this case.

This timing creates a great risk of actual bias and a resulting appearance of partiality because, even assuming that a Facebook “friendship” does not denote the type of relationship traditionally associated with the term “friendship”, it is unquestionably evidence of some type of affirmative social connection. As explained above, two Facebook users may only become “friends” when one user accepts another user’s “friend” request. Carroll’s choice to send a “friend” request to Judge Bitney, combined with Judge Bitney’s choice to accept that request before issuing his decision, conveys the impression that Carroll was in a special position to influence Judge Bitney’s ultimate decision — a position not available to individuals that he had not “friended”, such as Miller [the father].

Second, the great risk of actual bias and resulting appearance of partiality created by the Facebook connection between Carroll and Judge Bitney is heightened because the connection was not disclosed to any of the other parties or attorneys involved in the case.

… (Although we need not determine whether a bright-line rule prohibiting the judicial use of [electronic social media] is appropriate or necessary, we conclude that the circuit court’s undisclosed [electronic social media] connection with a current litigant in this case created a great risk of actual bias, resulting in the appearance of partiality.)

And yes, in case you were wondering, there are judicial ethical rules in Wisconsin prohibiting ex parte or undisclosed communications between a judge and one of the litigants in a case.

In State of New Mexico v Thomas,[27] during a murder trial that he was hearing, a judge made posts about the case to the Facebook page created for his re-election. For example, the judge posted “I am on the third day of presiding over my ‘first’ first-degree murder trial as a judge”, and afterwards “In the trial I presided over, the jury returned guilty verdicts for first-degree murder and kidnapping just after lunch. Justice was served. Thank you for your prayers”.

The appeal was decided on grounds other than any apprehension of bias. Nonetheless, the New Mexico Court of Appeals took the opportunity to warn judges about social media use in terms equally apposite to tribunal members, saying:[28]

we caution that “friending”, online postings, and other activity can easily be misconstrued and create an appearance of impropriety. Online comments are public comments, and a connection via an online social network is a visible relationship, regardless of the strength of the personal connection …

Judges should make use of privacy settings to protect their online presence but should also consider any statement posted online to be a public statement and take care to limit such actions accordingly …

The use of electronic social media also may present some unfamiliar concerns, such as the inability to retrieve or truly delete any message once posted, the public perception that “friendships” exist between people who are not actually acquainted, and the ease with which communications may be reproduced and widely disseminated to those other than their intended recipient.

Judicial officers and tribunal members will inevitably be personal friends with lawyers and barristers who they have worked with prior and subsequent to their appointment. The current generation of appointees uses, and increasingly will use, social media to stay in touch with their families and friends. This will mean that more and more judges and tribunal members will have online connections to those who appear before them.

Of course, not every social media connection will justify a claim of apprehended bias. A majority of the Supreme Court of Florida held recently that a Facebook friendship between a trial judge and a lawyer appearing before him, by itself, did not constitute sufficient grounds for the judge to recuse himself.[29] An unavoidable consequence of one’s appointment to an office, such as that of a judge or tribunal member, will be that one’s professional friends and acquaintances will appear before you as representatives of parties in matters that you must decide. That circumstance, of itself, cannot create a reasonable apprehension that the judge or tribunal member might not bring an impartial mind to the resolution of the dispute between the parties, one or more of whom is represented by your friend of acquaintance. The position may be different, as occurred recently in New Zealand, if the judge or member and the representative also have a business relationship together.[30]

However, things can become complicated when all the world can see what the friends or acquaintances say online about their relationship or thoughts. Hence, the need for care before posting or engaging on social media. In the Florida case, the court engaged in a detailed examination of the nature of traditional and Facebook friendships. Their Honors acknowledged that a Facebook friendship is different to the traditional variety, and does not give rise to inferences about whether or not there might be an apprehension of bias. Chief Justice Canady, giving the majority’s reasons, said that Facebook “friends” may be strangers, animals or inanimate objects. No doubt the Sydney Opera House or your local café will allow you to “friend” them on Facebook, but, that relationship defies any intelligible characterisation as a friendship in pre-21st century parlance. Chief Justice Canady sagely observed:[31]

A Facebook “friend” may or may not be a “friend” in the traditional sense of the word. But Facebook “friendship” is not — as a categorical matter — the functional equivalent of traditional “friendship”. The establishment of a Facebook “friendship” does not objectively signal the existence of the affection and esteem involved in a traditional “friendship”. Today it is commonly understood that Facebook “friendship” exists on an even broader spectrum than traditional “friendship”. Traditional “friendship” varies in degree from greatest intimacy to casual acquaintance; Facebook “friendship” varies in degree from greatest intimacy to “virtual stranger” or “complete stranger”.

However, even in that case there was a dissent. Justice Pariente said that even though an online friendship might convey no more intimacy than “complete stranger”, it still “may undermine confidence in the judge’s neutrality”.[32] This is because, first, a Facebook friendship gives access to the friend’s personal information, “including photographs, status updates, likes, dislikes, [and] work information …” on an almost daily basis. Secondly, the mere existence of a Facebook friendship, regardless of whether it is genuine, might convey to others that a lawyer is in a special position to influence that judge.

Judges associates and social media

The impact of social media engagement is not limited to the online conduct of judicial officers and tribunal members themselves. Thus, they must also be aware that the online activity of their staff and family can add to the perils of social media use.

Personal staff, especially young graduates or professionals such as a judge’s associate, or a tribunal member’s assistant or clerk, will be enthusiastic and no doubt often proud of the importance of their position and a matter that the judge or member is hearing. There will be temptations to post a “selfie” of how the associate or clerk is working, perhaps showing, inadvertently, a confidential document, or part of a draft of the judge’s or member’s decision. This can compromise the integrity of the court’s or tribunal’s process and public confidence in it.

In addition, the staff member actually has a life of his or her own, and can hold and express posts or opinions online about all manner of things that may concern issues that the judge or member may need to consider in a particular matter. Enterprising, misinformed or malign litigants or parties may perceive or seek to create some connection between the online posts or opinions of the staff member with those of the judge or member, merely because of the fact that the staff member works for him or her. Ordinarily, such an assertion could not rationally or reasonably give rise to an apprehension that the judge or member might be biased. However, that common sense position may not be enough as an unfortunate tipstaff to Harrison J found out in Gaynor v Local Court.[33]

There, a party who regularly advocated online in support for “traditional” heterosexual values, including opposition to same sex marriage, applied, by his counsel, for the judge to recuse himself. This was because the judge’s tipstaff had a large social media presence in which he openly supported the freedom of persons to identify as LGBTIQ+, and to participate in daily life without prejudice.

The tipstaff had sent an email to the parties enquiring, on behalf of the judge, about whether there were any new submissions for the upcoming hearing that had not reached the court file.

The party or his lawyers had accessed the tipstaff’s Facebook profile and discovered that, among other things, he had made a post in support of a University production entitled “Peter Pansexual”, which was reviewed as a “diverse, nuanced portrayal of the many shades of sexuality”; was “Facebook friends” with another judge’s tipstaff who had made a number of posts in support of gay and queer rights, including restricting the ability of religious schools to discriminate against people who identified as queer; and had participated in “Wear it Purple Day”, an annual day intended to foster awareness for lesbian, gay, bisexual, transsexual, intersexual and queer rights.

The party submitted that this gave rise to an apprehension of bias because of the personal process engaged in by a judge in hiring their staff. The party submitted that, somehow, “by innuendo”,[34] this suggested that the judge shared, or had some sympathy for these views. Justice Harrison dismissed this somewhat arcane reasoning process, saying:[35]

I am in any event unable to understand the relationship, if any, between the request made by my tipstaff with my authority in those terms and any apprehension of bias … The personal views of my tipstaves are largely unknown to me, except to the extent that they are revealed in the context of the relationship I have with them as my assistant in chambers … In my limited experience, cases are decided by judges, not their staff.

The social media habits of my own associates (one of whom has an unduly confident view of things in his draft of this speech) are unknown to me. I would hope they have the common sense to maintain sufficient privacy settings to avoid members of the public, like me, accessing their personal posts. Moreover, the Federal Court (like most modern workplaces) has policies as to the use of social media by its staff. But no such policies can affect what a judge’s staff has done before they are hired.

In any event, as Harrison J noted, ordinarily the staff member’s views make no difference to the duty or ability of a judge to decide cases fairly and impartially, both under the legal test for apprehension of bias or under a common sense idea of how cases are decided by courts. The position is likely to be the same for tribunal members who properly maintain their independence from all irrelevant considerations in arriving at their decision.

Self-deleting social media

One solution to the perils of life to which we are now exposed may be for social media to “expire”, hopefully, for those with guilty consciences at least, leaving no trace of one’s social media presence. This approach was first popularised by Snapchat, a social media platform used for sending photos with special effects and text on them, that permanently expire within 10 seconds of the recipient opening them. In the 1960s, the television program, “Mission Impossible”, portrayed the heroes receiving their mission through a taped message that literally said “this message will self destruct in 10 seconds” — an early forecast of destructibility.

Today, a variety of social media and online messaging platforms offer the ability to create posts and send messages which delete themselves after they are viewed, or after a set period of time. This has come about as individuals become increasingly concerned about their online “footprint” — the history of posts traceable to them. Instagram “stories” are public to all of a person’s followers for 24 hours after they are published. Facebook, Whatsapp and Telegram offer “secret conversations” that can last for as long or short a time as the sender chooses, and are then permanently deleted.

Even Gmail, the email platform hosted by Google, has begun offering a “Confidential Mode”, which allows users to send self-destructing emails. Once received, the contents of these emails cannot be copy-pasted, downloaded, printed or forwarded.[36]

It may be that none of this self-destruction will necessarily prevent people creating a screenshot or photographing the contents of temporary or private posts and messages. In this year’s federal election campaign, a slew of candidates resigned after some of their old, perhaps forgotten or deleted, social media posts came back to life. At least one of those candidates had deleted the posts prior to the campaign, but they resurfaced, probably because someone had taken screenshots earlier while they were public.[37] There are also tools online that can uncover deleted tweets, long after those have been deleted by the person responsible for them.[38] The lesson is that attempts at online self-effacement do not guarantee against later self-destruction.

Avoid social media altogether?

The appropriate course may still be not to have social media at all. Judge Catherine Shaffer, the 2018 president of the American Judges’ Association, recently acknowledged that judges can decide for themselves whether to use social media. However, she recommended they “steer clear” because “misperception is all too easy” and “it is extraordinarily difficult to prevent improper ex parte contacts”.[39]

Similarly, the Council of Chief Justices of Australia and New Zealand’s Guide to Judicial Conduct[40] cautions that although “there is no reason in principle to deny judges the use of social media”, “a judge should be aware of the risks that go with the use of social media, and should act with care in light of these risks”.

Alongside the risks already mentioned today, the Guide notes that while it is important that judges understand and make use of the strictest privacy settings available, they should not rely on this. That is because others can widely disseminate material that the judge wishes to be kept private, and those privacy settings also can change.

Family members of judicial officers and tribunal members should also be aware of the risks that social media presence can pose to their partner, parent, child or sibling’s work. First, there are issues as to the personal security of the judge or member. Secondly, like the tipstaff mentioned above, these actions online should not, alone, give rise to an apprehension of bias. But it is always better to be safe than sorry, and better that courts and tribunals’ time is not wasted with such applications at all. Family members should recognise these risks and know to alert judicial officers and tribunal members if a case before them is linked in some way to their online network or presence.

Part III: Social media and mass media in courts and tribunals

Ordinarily, proceedings in court are open to the public and the public have a legal right to be present and to report what occurs in such proceedings, and to criticise it: John Fairfax & Sons Ltd v Police Tribunal of NSW.[41] Except as required by statute, a court can only restrict access to, and reporting of, court proceedings where it is necessary to do so in the interests of justice. The same very fundamental common law principles do not apply to administrative proceedings, which traditionally do not require transparency or public scrutiny.

Judicial officers and tribunal members must be conscious of what uses of social media may be appropriate during a hearing when persons in the court or room can use online platforms to disseminate or discuss the subject matter of the hearing. This is unavoidable. Regardless of whether the judge or tribunal member has chosen whether or not to use social media for his or her own personal purposes, it will be important, from time to time, to make decisions about how to manage the flow of instantaneous information to protect the integrity of the proceeding. Such a flow can be from journalists, litigants, bloggers, lawyers, part representatives or a member of the public.

It is increasingly common for journalists in the mass media to “live tweet” while sitting in open court. This is a way of immediately updating their audience, online, on developments in the courtroom or tribunal. Although most courts and tribunals have the power to control the use of electronic devices, their use has become more frequent and now it is not uncommon to see journalists, lawyers and others on phones or even laptops in court.

Whatever control a judge or tribunal member can exert over online transmission directly from the court or hearing room, that control has become less and less effective when the publisher communicates from outside the hearing.


Speaking the right social media language means recognising that the role of courts and tribunals is not (and never was) confined to the walls of the courtroom or tribunal. Those hearing and determining proceedings today must be aware, more than ever, of the risks to justice which can arise from social media.

In part, this means regulating the behaviour of parties and observers in court or a hearing. But it also requires judicial officers and tribunal members to be aware of their own conduct online. It is too late to forbid judicial officers and tribunal members from using social media themselves, even if that were a good idea. The vast majority of future appointees will already have social media accounts, and as I have mentioned, the details of those may be impossible to ever truly delete.

There is also, of course, no way to stop members of the public from using social media to comment on court and tribunal processes. To the extent there are ways to limit this, they should be employed at least by judicial officers and courts with reticence, given the fundamental role of open justice in our society.

[1] Paper presented at the Council of Australian Tribunals National Conference, Melbourne Convention and Exhibition Centre on 6 June 2019. Published by the Federal Judicial Scholarship database on Austlii at (FCA) [2019] FedJSchol 6.

[2] A judge of the Federal Court of Australia and an additional judge of the Supreme Court of the ACT and the Supreme Court of Norfolk Island. The author acknowledges the assistance of his associate, Ryan Hunter, in the preparation of this paper.

[3] F Gibb, “Baroness Hale to lay down the law on MasterChef”, The Times, 30 March 2018 at www.thetimes.co.uk/article/baroness-hale-to-lay-down-the-law-on-masterchef-df3pbtfz7, accessed 3 June 2021.

[4] J Biskupic, “Justice Ruth Bader Ginsburg calls Trump a ‘faker’, he says she should resign”, CNN, 13 July 2016 at https://edition.cnn.com/2016/07/12/politics/justice-ruth-bader-ginsburg-donald-trump-faker/index.html, accessed 3 June 2021.

[5] D Scarinci, “Justice Ginsburg is social media darling”, Observer, 2 July 2015 at https://observer.com/2015/07/justice-ginsburg-is-social-media-darling/, accessed 3 June 2021.

[6] Thomas v Bradbury, Agnew & Co Ltd [1906] 2 KB 627 at 628.

[7] B Salt, “Online warriors bum me out”, The Weekend Australian Magazine, 3 August 2019.

[8] S Kemp, “Digital 2019: Global internet use accelerates”, We Are Social, 30 January 2019 at https://wearesocial.com/blog/2019/01/digital-2019-global-internet-use-accelerates, accessed 3 June 2021.

[9] D Cowling, “Social Media Statistics Australia — December 2018”, SocialMediaNews, 1 January 2019 at www.socialmedianews.com.au/social-media-statistics-australia-december-2018/, accessed 3 June 2021. [Note: in May 2021, there were 17 million Australians using FaceBook each month; 9 million Instagram accounts and 5.8 million Twitter accounts.]

[10] Cowling, ibid. [Note: in May 2021, 16.5 million Australians watch YouTube each month, and 2.9 million use WeChat.]

[11] Re Refugee Review Tribunal; Ex p H [2001] HCA 28 at [28]–[29] per Gleeson CJ, Gaudron and Gummow JJ; NADH of 2001 v Minister for Immigration and Multicultural and Indigenous Affairs (2004) 214 ALR 264 at [14]–[21] per Allsop J, with whom Moore and Tamberlin JJ agreed. (In the following discussion I have drawn heavily on what Jagot J and I wrote in Minister for Immigration and Citizenship v SZQHH (2012) 200 FCR 223 at [37]–[45].

[12] See, for example, in relation to judges: Re JRL; Ex p CJL (1986) 161 CLR 342 at 352 per Mason J.

[13] (1969) 122 CLR 546 at 554.

[14] (2001) 205 CLR 507 at [72].

[15] ibid at [98]–[100].

[16] ibid at [179]–[192].

[17] ibid at [180].

[18] ibid.

[19] ibid at [185]–[186]. [Bold emphasis added.]

[20] R v London County Council; Re Empire Theatre (1894) 71 LT 638 at 639, per Charles J.

[21] See eg, Johnson v Johnson (2000) 201 CLR 488 at [13].

[22] Minister for Immigration and Multicultural Affairs v Jia Legeng (2001) 205 CLR 507 at [187] per Hayne J.

[23] ibid at [192].

[24] See the account in J Ronson, So you've been publicly shamed, Riverhead Books, 2015.

[25] In re the paternity of B J M: Timothy W Miller v Angela L Carroll [2019] WI App 10.

[26] ibid at [21]. [Bold emphasis added.]

[27] State of New Mexico v Thomas 376 P 3d 184 (NM, 2016) at [8].

[28] ibid at [49].

[29] Law Offices of Herssein and Herssein v United Services Automobile Association SC17-1848 (Fl, 2018).

[30] See Saxmere Company Ltd v Wool Board Disestablishment Co Ltd [2010] 1 NZLR 35.

[31] Law Offices of Herssein and Herssein v United Services Automobile Association SC17-1848 (Fl, 2018) at 13.

[32] ibid at 23.

[33] [2019] NSWSC 516.

[34] ibid at [36].

[35] ibid at [33]–[35].

[36] T Haselton, “How to send self-destructing emails in Gmail”, CNBC, 20 August 2018 at www.cnbc.com/2018/08/20/how-to-send-self-destructing-email-gmail-confidential-mode.html, accessed 3 June 2021.

[37] “Social media minefield for election candidates”, InDaily, 3 May 2019, https://indaily.com.au/news/2019/05/03/social-media-minefield-for-election-candidates/, accessed 8 June 2021.

[38] K Hodgkins, “Undetweetable lets you read deleted tweets”, Gizmodo, 3 August 2011, www.gizmodo.com.au/2011/08/undetweetable-lets-you-read-deleted-tweets/, accessed 8 June 2021.

[39] “Are Facebook friends really friends?”, National Center for State Courts, 29 August 2018, https://content.govdelivery.com/accounts/USNCSC/bulletins/209842b, accessed 8 June 2021; see Law Offices of Herssein & Herssein, PA v United Services Automobile Association, Case No SC17-1848 at 22.

[40] The Australasian Institute of Judicial Administration, Guide to Judicial Conduct, 3rd edn (rev), 2022 at p 43.

[41] (1986) 5 NSWLR 465.